Hypothesis test problem




I was trying to use ANOVA to test if there is significant difference between employee engagement score in different employee tenure groups.But I found out that Distribution of my dependent variable (i.e Engagement score ) is not normal,which violates the assumption of ANOVA.

So, I used non-parametric alternative of ANOVA which is Kruskall-Wallis test which doesn’t assume normality but again it is violating its assumption of Homogeneity of variances as bartlett test’s p-values is significant i.e less than 0.05.

Can you guide what should I do?

Looking forward for your response!


Is there any way to share your data? From what I read it sounds like you want to do a One-Way ANOVA: Randomized Block Design.